Aprovação no Comitê de Ética em Pesquisa

7. Cronograma

Atividade Mês (2012/ 2013)

Mar Abr Mai Jun Jul Ago Set Out Nov Dez Jan Fev Mar Abr Mai Jun Jul Ago Set Out nov Dez Levantamento Bibliográfico Aquisição de materiais Levantamento de casos Realização das reações de Imunohistoquimica Análise e contagem das células Tabulação e Análise de dados Finalização do Artigo e defesa Monografia

O presente relatório é parte do estudo realizado em Pelotas, RS com o título ―Avaliação da expressão de proteínas reguladoras do metabolismo ósseo na periodontite crônica em fumantes e não-fumantes‖, ocorrido de agosto de 2010 à fevereiro de 2014 no Laboratório de Histopatologia da Faculdade de Odontologia da Universidade Federal de Pelotas.

A aquisição dos anticorpos foi realizada previamente e após a qualificação do projeto de dissertação, porém algumas alíquotas só chegaram no laboratório no final de novembro de 2013, e assim realizamos os testes e otimizações até o inicio de janeiro de 2014 onde seguimos com as reações para imunomarcação das amostras.

Para quantificar a expressão dos anticorpos nos tecidos, foi acordado que não poderíamos realizar dupla marcação, conforme proposto inicialmente, em virtude das Nod1 e Nod2 serem marcadores de citoplasma inviabilizando desta forma a distinção de cada um nos tecidos analisados. Devido à necessidade de repetição das reações e o tamanho pequeno dos amostras, a leitura das Nod em terços em todas as lâminas ficou prejudicado. Assim, nas lâminas marcadas para Nod optamos então por fazer contagem das células inflamatórias totais por campo em cada uma das amostras.

Os resultados desta pesquisa serão divulgados das seguintes maneiras: Dissertação de Mestrado, conforme normas do Programa de Pós Graduação em Odontologia da UFPel; artigo científico a ser publicado em revista internacional ―Journal of Clinical Periodontology‖ (normas de publicação no anexo 1).

Title: Increased expression of NOD, RANKL, OPG and MMP in smokers and nonsmokers with periodontitis

Aline Ferreira Almeida1, Luis Eduardo Rilling da Nova Cruz1, João Batista César Neto2, Ricardo Alves de Mesquita3, Adriana Etges1, Fábio Renato Manzolli Leite1

1

Graduate Program in Dentistry, School of Dentistry, Federal University of Pelotas, Pelotas, RS, Brazil.

2

Department of Stomatology, School of Dentistry, University of São Paulo, São Paulo, SP, Brazil.

3

Department of Oral Surgery and Pathology, School of Dentistry, Federal University of Minas Gerais, MG, Belo Horizonte, Brazil.

Running title: Nod, RANKL, OPG, MMP in periodontitis

Correspondence Author:

Fábio Renato Manzolli Leite

457, Gonçalves Chaves St. room 511 Pelotas/RS – Brazil

96015-560

Aim: Periodontitis is a chronic disease modulated by cigarette consumption inducing innate

and adaptive immune responses within the periodontal tissues. To assess these immune responses, we evaluated the expression of nucleotide-binding oligomerization domains (Nod)- 1 and -2, receptor activator of nuclear factor kappa-B ligand (RANKL), osteoprotegerin (OPG), CD20 (B-cells), CD68 (monocytes/macrophages), CD45RO (T-cells) and matrix metalloproteinases (MMP)-8 and -9 in periodontal tissues of smokers and nonsmokers in order to identify possible differences.

Methods: Gingival tissues were obtained from periodontally health subjects (n=10), and

patients with chronic periodontitis smokers (n=15) and nonsmokers (n=16). Nod1, Nod2, RANKL, OPG, CD20, CD68 and CD45RO expression was determined by immunohistochemistry (IHC). MMP-8 and -9 activity were verified by zymography.

Results: Periodontitis-affected patients presented higher levels of all studied proteins when

compared with healthy sites (P < 0.001). However, no differences were observed between smokers and non-smokers (P > 0.05). All proteins were more expressed at the apical portion of periodontal pocket compared with the coronal portion.

Conclusions: Considering periodontitis as an infectious disease, its progression induce the

migration of B- and T-cells and monocytes/macrophages to inflamed site. In addition, pattern recognition receptors and bone turnover regulators are overexpressed. No differences were observed between smokers and nonsmokers.

Keywords: Periodontal Diseases; Nod; Rankl; Osteoprotegerin; Matrix metalloproteinases

Clinical Relevance

Scientific Relevance for the study: No literature exists considering the expression of Nod-like

receptors, bone turnover regulators and immune cells distribution along the periodontal pocket in smokers and nonsmokers.

progression between smokers and nonsmokers are not associated with inflammatory cells and gelatinases expression.

Practical Implications: Since no differences in cell population, bone remodeling and matrix

degrading proteins were observed, smoking patients should be advised to reduce or quit the habit and enroll into a more constant supportive periodontal therapy to control disease progression.

Introduction

Bacterial infections are detected by the innate immune system through transmembrane and cytosolic receptors (Chamaillard et al., 2003). It is well known that Toll-like receptor (TLR)-4 recognizes gram-negative microorganisms eliciting a complex intracellular signaling cascade resulting in proinflammatory mediators secretion, such as interleukin (IL)-1, IL-6, tumor necrosis factor alpha, among others (Garcia de Aquino et al., 2009). More recently, two members of a family of proteins called nucleotide-binding oligomerization domain (Nod)-1 and -2 were found to recognize substructures of bacterial peptidoglicans (Inohara et al., 2003, Girardin et al., 2001).

Gram-negative and some species of Gram-positive bacteria produce the amino acid meso-diaminopimelic acid (meso-DAP) responsible for Nod1 activation (Girardin et al., 2001). On the other hand, all Gram-negative and -positive bacteria present muramyl-dipeptide (MDP) which is recognized by Nod2 (Kim et al., 2008). Mutations in NOD2 gene cause hyperinflammatory responses to bacteria components such as in Crohn’s disease and Blau´s syndrome, both autoinflammatory diseases (Hugot et al., 2001). In cases of bone resorption, osteoblast stimulated by MDP (Nod2 agonist) synergistically enhances osteoclast formation induced by lipopolysaccharide (LPS), IL-1α, and tumor necrosis factor alpha (TNF-α) through

2005).

Destruction seen in animal and human with periodontitis is potentiated by smoking habits (Cesar-Neto et al., 2006). Nicotine combined with LPS increased the release of prostaglandin E2 (PGE2), IL-1β and TNF-α in circulating monocytes (Payne et al., 1996, Ryder, 2007) and IL-6 in fibroblasts and osteoblasts (Wendell and Stein, 2001). Wistar rats with ligature-induced periodontitis exposed to cigarette smoke presented higher levels of bone loss, matrix metalloproteinase 2 (MMP-2), IL-6 and interferon gamma (IFN-γ) (Cesar-Neto et al., 2006).

Smoking patients with periodontitis tended to present lower levels of the receptor activator of nuclear factor kappa-B (RANK) decoy osteoprotegerin (OPG) compared to nonsmoking periodontitis patients in blood stream (Lappin et al., 2007). In addition, RANKL:OPG ratio was found higher in smokers possibly explaining the higher attachment loss observed in smokers.

However, some authors failed to observe in vitro a tendency of increase in inflammatory genes expression in smoking periodontal patients after smoking cessation (Morozumi et al., 2004, Ouyang et al., 2000). It is important to emphasize that most of the studies that evaluated periodontal disease pathogenesis associated with smoking habits evaluated crevicular fluid samples, cell cultures or processed tissues.

Thus, this study evaluated differences in RANKL and OPG immunohistochemical detection in periodontal pockets from smoking and not smoking patients with periodontitis. In addition, Nod1 and Nod2 expression was evaluated in healthy and periodontitis-affected gingival tissues.

The study was approved by the Federal University of Pelotas institutional ethical committee (#038/2008).

All subjects were submitted to anamnesis and to periodontal and radiographic examination. Supragingival calculus was removed to allow periodontal probing. Patients were categorized according to the classification of the American Academy of Periodontology (Armitage, 1999) into healthy and chronic periodontitis groups.

Inclusion criteria included dentate patients (at least 14 natural teeth excluding third molars), systemically healthy with no evidence of systemic periodontal modifiers other than smoking (diabetes mellitus, osteoporosis, and medications known to influence periodontal tissues) (Garlet et al., 2003). Smokers should use at least 20 cigarettes/day for at least 5 years and nonsmokers should never had smoked.

Exclusion criteria included patients with systemic modifiers of periodontal disease as cited above; who had taken systemic antibiotic, anti-inflammatory, or other drug therapy in the last 3 months; who had received previous periodontal therapy in the last 2 years or pregnant/lactating women (Garlet et al., 2003).

Chronic periodontitis subjects presented moderate to advanced chronic periodontitis (>1 tooth per sextant with probing depth > 5 mm and attachment loss > 3 mm, and extensive radiographic bone loss) (Garlet et al., 2003). Gingival biopsy sites in healthy control subjects did not exhibit radiographic bone destruction, as well as having clinical probing depths less than 3 mm without sulcular bleeding on probing.

Gingival biopsies

Samples were surgically obtained from subjects with teeth to be extracted, root scaled, or having their crown lengthened for restoration. Gingival tissues were obtained from

and nonsmokers (n=16). Probing depth (PD), gingival margin level (GM) and bleeding on probing (BP) were recorded (Garlet et al., 2003).

Biopsies were obtained of the proximal face under the contact point using intrasulcular incisions. In case of isolated teeth two parallel 5mm incisions connected by a perpendicular incision (fig. 1 and 2).

Figure 1. Representation of sample removal for an isolated tooth. Oclusal view with incision scheme and the radiographic image representing the depth of the incision.

Figure 2. Operative procedure for proximal surface under contact point. A: Bone defect between the premolars; B: Intrasulcular incisions delimiting the area. C: Full thickness flap; D: Immediate post-operative.

Gingival tissue were fixed in 10% buffered formalin, serially cut in the coronal plane, and embedded in paraffin for sectioning. The formalin-fixed, paraffin-embedded tissue specimens were sectioned at 5 μm. Sections were mounted on silanized slides, deparaffinized in xylene, blocked in 1:1 H2O2 in absolute methanol, and processed for antigen retrieval.

The sections were reacted with specific antibodies as shown in Table 1. After being washed with TRIS-HCl, the sections were incubated with biotinylated immunoglobulins (Universal Ab, Dako Corporation, CA, USA) for 20 min at room temperature and washed with TRIS-HCl to remove any unreacted antibodies. The sections were then treated with peroxidase-conjugated streptavidin (DAKO) for 10 min, and washed and reacted with DAB (3,3-diaminobenzidine tetrahydrochloride; DAKO) in the presence of 3% H2O2 to develop

color. The sections were counterstained with Mayer´s hematoxylin and mounted with Entellan (Merck; Darmstadt, Germany).

Table 1. Antibodies and parameters used for immunohistochemistry reactions

ANTIBODY CLONE BRAND HOST DILUTION INCUBATION

CD20 7D1 Novocastra Mouse 1:200 60’

CD68 PG-M1 Dako Mouse 1:50 60’

CD45RO UCHL1 Dako Mouse 1:100 60’

RANKL N-19 Santa Cruz Goat 1:100 60’

OPG N-20 Santa Cruz Goat 1:50 60’

NOD1 bs-7085R Bioss Rabbit 1:800 12h

NOD2 bs-7084R Bioss Rabbit 1:800 12h

Novocastra laboratories Ltd, Newcastle, UK; Dako Cytomation, Glostrup, DN; Santa Cruz Biotechnology, Santa Cruz, CA, USA; Bioss, Woburn, MA, USA

Slide evaluation was independently performed by two blinded examiners. In case of disagreement consensus was reached by discussion with a pathologist. For RANKL, OPG, CD20 (B-cells), CD68 (monocytes/macrophages) and CD45RO (T-cells) positive cells were

such a way (magnification factor 200) that one square of the counting grid corresponded to 0.0025 (mm2) (Garcia de Aquino et al., 2009). Sample was divided in coronal, medium and apical thirds (fig. 3). Cells expressing Nod1 and Nod2 were counted with the grid by the hot spot technique (Shimura et al., 2000).

Figure 3. Illustration of an immunohistochemistry reaction divided in coronal, medium and apical thirds. # tissue in direct contact with tooth; * area of the connective tissue where cells were counted. Magnification 100X.

Gelatin zymography

MMP-2 and MMP-9 activity were assessed by zymography. Part of the gingival tissue were dissected immediately following excision, pooled and washed (at 24°C for 30 min) in Dulbecco’s modified Eagle’s medium (DMEM) containing 80 mg ml−1 of gentamicin. After 24h the supernatant was frozen at −70°C until analysis for enzyme activity. Proteolytic activity was examined on 10% polyacrylamide gels containing 0.05% gelatin. The conditioned medium was mixed with an equal volume of non-reducing sample buffer [2% sodium dodecyl sulfate (SDS); 125 mM Tris–HCl (pH 6.8), 10% glycerol, and 0.001% Bromophenol Blue) and then electrophoresed. Gels were washed in 2% Triton X-100 for 60 min at room temperature and then incubated at 37°C for 24 h in 50 mM Tris–HCl buffer,

G-250 (Bio-Rad, Richmond, CA, USA). The gelatinolytic activity was detected as unstained bands. To assess the identity of the lytic bands present in the conditioned media, parallel experiments using proteinase inhibitions were performed. Gelatin-containing gels were incubated in Tris–CaCl2 buffer at 37°C for 24 h with the addition of 0.5 mM EDTA (Reagen,

São Paulo, SP, Brazil) to inhibit lytic activities caused by MMP, while 0.5 mM N-ethyl- maleimide (NEM) was used to inhibit activities caused by serine proteinases. MMP-2 and -9 were identified by comparing the molecular weight (Perfect Protein™ AP Kit, Novagen, Darmstadt, Germany). Images were analyzed densitometrically after digital imaging capture (Image Quant 100 – GE Healthcare), using ImageJ 1.45 software (Wayne Rasband, NIH, USA).

Scoring System and Statistics

Statistical significance was analyzed with t-test for data with normal distribution and Mann-Whitney followed by Dunn test for non-parametric; P < 0.05 were considered significant. The spearman coefficient was used to study correlations between continuous variables. A statistical software was used to perform the statistical analysis (GraphPad Prism 5.0 Software, San Diego, CA, USA).

Results

Forty-one patients, which included 20 men participated in this study. No statistical differences were observed for age and gender among the three groups, and for pocket depth (PD) and clinical attachment loss (CAL) between the two periodontitis groups. Table 2 shows the sample distribution into the groups according to demographic and clinical variables.

group (mean values and standard deviation) Healthy Periodontitis Nonsmoking Periodontitis Smoking Age (years) 39.8 (±5.4) 42.3 (±6.7) 46.4 (±7.7) Gender F= 5/10 M=5/10 F= 8/15 M= 7/15 F= 7/15 M=8/15 Pocket depth (mm) 2.1 (±0.8) 7.8 (±1.4) 7.9 (±1.2) Attachment loss (mm) 0 7.6 (1.6) 7.3 (2.2) Cigarettes smoked/day - - 17 (4.1) Years of smoking - - 22.9 (12.9)

A significant statistical difference was observed between the healthy and both periodontitis affected groups for all IHC and gelatin zymography reactions (P < 0.001). However, no differences were seen between smokers and nonsmokers (Table 3).

Table 3. Cell counting for each antibody marker by immunohistochemistry in smokers and

nonsmokers affected by periodontitis IHC parameter

Nonsmokers Positive cells (mean ±

SD)

Smokers Positive cells (mean ±

SD) P value CD20 173.25  330.35 162.27  171.20 0.745 CD45RO 614.31  901.82 586.60  626.38 0.887 CD68 155.10  182.78 145.53  106.40 0.174 Total (CD20 + CD68 + CD45RO) 942.56  1196.11 894.40  765.02 0.267 OPG 193.31  89.36 142.31  72.33 0.494 RANKL 566.59  277.36 614.30  270.58 0.824 NOD1 423.60  188.26 531.15  222.76 0.247 NOD2 346.54  134.25 370.19  179.01 0.564

coronal third of the periodontal pocket for both groups (Table 4). Healthy samples were not split in thirds since PD was very short to be divided and the inflammatory cells scarce.

Table 4. Cell counting along the periodontal pocket split in thirds for each antibody marker by

immunohistochemistry in smokers and nonsmokers affected by periodontitis. Labelling Sample region

(thirds) Nonsmokers Positive cells (mean ± SD) Smokers Positive cells (mean ± SD) CD20 Coronal 21.31  21.57a 22.40  38.10a Average 56.50  88.12ab 58.46  70.25ab Apical 95.43  249.67b 81.40  89.98b CD45RO Coronal 54.93  85.84a 71.27 70.01a Average 227.25  335.56ab 160.06  167.18ab Apical 332.12  549.18b 335.26  500.17b CD68 Coronal 53.37  89.73a 25.2  23.81a Average 62.81  94.98a 60.00  61.82b Apical 38.81  40.20a 60.33  54.32ab OPG Coronal 21.31  11.57a 22.40  12.10a Average 76.57  28.12b 58.46  20.25ab Apical 95.43  49.67b 61.45  39.98b RANKL Coronal 71.27 70.01a 54.93  15.84a Average 160.06  107.18ab 227.25  105.56b Apical 335.26  100.17b 332.12  149.18b *Different lowercase letters indicate statistical significant difference

A statistically significant difference was observed between the number of cells positive for Nod1 and Nod2 in healthy and periodontitis affected samples (P < 0.001). However, no differences were observed in Nod1 and Nod2 expression between smokers and nonsmokers (P = 0.564 and P = 0.856, respectively).

Illustrative images of the reactions in the middle third of the samples for smoking patients are presented in Figure 4.

Figure 4 - Increased in immunoreactive cells number in chronic periodontitis. Periodontal tissue sections were submitted to immunohistochemical reaction for CD20 (A), CD45RO (B), CD68 (C), RANKL (D), Nod1 (E), Nod2 (F). Illustrative image of the middle third section of the smoking group. Magnification 400Xnão é 40x?.

In addition, no differences were seen in MMP-2 and MMP-9 expression (P = 0.951 and 0.183, respectively) between smokers and nonsmokers with periodontitis. Moreover, gelatinases activity were not different between smokers and nonsmokers (P = 0.671).

Discussion

The present study was the first to analyze by IHC biopsies from smoking and nonsmoking patients with periodontitis in order to evaluate the distribution of inflammatory cells and markers along the periodontal pocket to justify a possible increase in CAL in smokers.

periodontitis development. However, no differences were seen on cell concentration or cell distribution along the periodontal pocket between the smoking and nonsmoking groups. Hypothesizing that smokers with periodontitis have higher CAL, previous studies in rats (Cesar-Neto et al., 2006, Cesar-Neto et al., 2007) showed that some cigarette components could stimulate a higher expression of inflammatory cytokines and that CAL was more associated with the mediators expressed than with cell population and concentration.

Lymphocytes play a major role in periodontitis pathogenesis (Khalaf and Bengtsson, 2012). The nature of adaptive immune response is controlled by T-cells, which regulate the B- cells population and immunoglobulin release. T-cells are responsible for the production of different mediators, such as interferon-gamma, IL-1, -6, RANKL, adhesion molecules and chemokines (Khalaf and Bengtsson, 2012). These mediators direct the immune response from gingivitis to periodontitis directly or indirectly stimulating bone resorption and CAL (Loos et al., 2004, Gemmell and Seymour, 2004).

Higher T-cell proliferation was observed in heavy smokers (Loos et al., 2004). Authors concluded that the increase in T-cells responsiveness and number would be one of many factors that explain why smoking is a risk factor for periodontitis. Our results failed to show an increased number of T-cells in smokers than nonsmokers. However, the present lymphocytes should be checked for their activity since an increase in cytokines release could induce RANKL expression.

Osteoclastogenesis is mediated by the interaction among three molecules RANK, RANKL and OPG. OPG was the first protein to be described of this metabolic axis as an osteoclast inhibitor (Tsuda et al., 1997, Simonet et al., 1997). A few months later, RANKL was reported as a tumor necrosis factor capable of stimulating dendritic cells to induce T-cells proliferation (Anderson et al., 1997, Wong et al., 1997). RANK was cloned from a dendritic cell cDNA library because when activated it extended the cell survival (Anderson et al.,

(Nakagawa et al., 1998). In sum, the RANK/RANKL/OPG axis acts by the interaction of RANKL-RANK, which positively regulates osteoclastogenesis and is counterbalanced by OPG, a false natural decoy for RANKL (Hofbauer and Heufelder, 2001).

According to our data, periodontitis affected patients presented higher levels of RANKL, OPG, Nod1 e Nod2 when compared with healthy ones, which is proportional to the levels of CD20, CD68, CD45RO positive cells. (Tang et al., 2009) corroborate the fact that RANKL levels are not affected by smoking, but OPG levels slight reduce causing bone resorption. For the first time, we show that RANKL and OPG levels are higher at the apical portion of the pocket, where a high gram-negative bacterial load is present and closer to alveolar bone.

Various stimuli can induce increased expression of RANKL specially in osteoblasts, such as parathyroid hormone, 1,25-dihydroxyvitamin D , interleukin- 1 and lipopolysaccharide (Gallagher, 2008). Intracellular signaling by TLRs or bacterial stimulation via IL-1R induced by interleukin-1 requires the activation of intracellular proteins like p38 MAPK and especially NF-kB (O'Neill and Bowie, 2007, Stone et al., 1988). The role of MyD88 adapter protein in the expression of proinflammatory genes induced via activation of TLRs involves the recruitment of signaling proteins common to both paths located upstream. It is already known that IRAKs (IL-1 receptor-associated kinase) activate TRAF6 (tumor necrosis factor receptor-associated factor 6-associated), which subsequently recruits TAK1 (Transforming growth factor β-activated-kinase-1) and TABs (TAK1 binding proteins). From this point two signaling pathways can be distinctly activate through the recruitment and activation of IKK (IκB kinase) complex and MAP3Ks (as MKK3 and MKK6, upstream activators of p38 MAPK) (Adhikari et al., 2007, Kobayashi et al., 2002).

adapter proteins, TRIF (TIR domain-containing adapter protein inducing IFN-b) and TRAM (-related adapter molecule) (Krishnan et al., 2007). Thus, the same intracellular signaling pathway through branching and plasticity regulate distinct genes in different cell types or, alternatively, in the distinct regulation of a single gene in different cell types.

It is speculated that Nod2 and probably Nod1 binds to the serine / threonine kinase RIP2 (also known as RICK or Cardiak), which is involved in signal transduction via NF-kB (Hasegawa et al., 2008). Results indicate that Nod1 and Nod2 signaling pathway is required for expression of RANKL induced by agonists of TLR2 and 4, but only Nod1 is required for expression of RANKL after stimulation by IL-1β (Hasegawa et al., 2008). In this study, RANKL, Nod1 and Nod2 levels were higher in periodontitis affected patients, in which alveolar bone resorption is present.

Nod proteins effect seems to be potentiated by the presence of IFN-γ and TNF-α (Hosokawa et al., 2010). Kawai et al. (2000) observed that Th1-type cells migrated to periodontitis sites induced by MDP. Takahashi et al. (2006) reported that mononuclear cells and fibroblasts expressed Nod1 and Nod2 in sites affected by periodontitis, as observed in this study, and Nod expression was enhanced by IFN-γ, TNF-α and LPS.

1.1.1. Importantly, the information concerning the role of Nod proteins as inflammatory signaling mediators involved in the innate immune response is still scarce. However, there is evidence highlighting their role in the expression of several inflammatory cytokines, including IL-1β, IL-6, IL-8 and TNF-α both in monocytic cells and epithelial cells (Uehara et al., 2007, Kim et al., 2004, Kobayashi et al., 2005) and modulation of cytokine expression via activation of TLRs (Netea et al., 2005).

Since a mutation in Nod2 protein is associated with Crohn´s disease in humans with a hyperinflammatory phenotype (Inohara et al., 2003) and considering that Nod higher expression is induced by proinflammatory cytokines and LPS (Gutierrez et al., 2002), Nod

periodontal pockets.

Acknowledgments

The authors are grateful to the Research Support Foundation of the State of Rio Grande do Sul for the